KR102528815B1 - Image-based displacement measurement apparatus related to pile driving - Google Patents

Abstract

The present invention relates to an image-based displacement measuring device related to pile driving, which includes two targets each attached to a hammer of a pile driver and a predetermined position of a pile and having different shapes so as to be distinguished; A camera installed toward the target, continuously acquiring video images to measure the displacement of the target, and including two targets having different shapes in one frame of the acquired video images, and video image data from the camera. By including a field terminal capable of receiving, calculating the displacement of the target, and giving operating instructions by the operator through wireless communication, it is possible to simultaneously measure the amount of penetration, rebound, and hammer fall with one image. , the precision of the measured displacement is very high by applying the unique correlation value calculation method and interpolation method without being affected by vibration by driving.

Description

Image-based displacement measurement apparatus related to pile driving}

The present invention relates to an image-based displacement measuring device related to pile driving, and more specifically, not only can measure the amount of penetration, rebound and hammer fall at the same time with one image, but also is not affected by vibration caused by driving. It relates to a displacement measuring device with very high precision of the displacement amount measured by applying a unique correlation value calculation method and interpolation method without

In general, in order to construct civil engineering works such as high-rise buildings, bridges, ports, etc., pile construction is performed on the ground prior to constructing structures, that is, driving piles in a vertical direction. These piles are driven into the ground using a pile driver to penetrate into the solid ground to resist the vertical force.

Such pile construction is a basic process of civil engineering, and it is becoming more important to calculate the depth at which piles are driven and the allowable bearing capacity for the safety of increasingly high-rise structures.

When driving such a pile, the amount of penetration of the pile is managed by measuring the displacement generated in the pile by driving, such as the amount of penetration and rebound. This is because not only can it be confirmed whether the pile foundation is properly constructed on the ground through the displacement behavior of the pile measured during driving, but also the allowable bearing capacity of the pile can be indirectly evaluated through an empirical method.

The allowable bearing capacity of these piles is highly dependent on complex site ground conditions or construction conditions, and since it is difficult to accurately predict the bearing capacity of piles during design, field tests are essential to confirm the bearing capacity of piles during construction. .

However, conventionally, in practice, the measurement of the displacement of the driving pile is mainly performed manually. Grid paper is attached to one side of the penetrating pile, and the operator directly holds a felt pen and stands right next to the pile to be driven, holding the pen on the graph paper. Penetration amount was measured by the touch-up method.

Such manual work has problems with the safety of the operator, limitations in the measurement range, errors caused by environmental factors, the possibility of arbitrary manipulation by the operator, and the need to analyze the data separately in the room after measuring the data in the field. there is

In order to solve this problem, there have been development cases in which various measurement and optical equipment, such as using lasers or multiple cameras, have been applied to the displacement measurement of driving piles. It has been around for a long time and has not been actively used yet.

Therefore, there is a need for a displacement measuring device related to pile driving that is precise through image analysis using a single reliable image collection unit and does not need to be close to the pile.

On the other hand, as the prior art, as disclosed in Republic of Korea Patent Registration No. 10-2200824 (published on January 12, 2021), a shooting target attached to a certain position of the pile and installed; a camera installed around the driving pile to acquire an image of the displacement amount of the shooting target generated according to the driving penetration amount of the driving pile; an acceleration sensor detachably installed on the shooting target to measure a rebound amount after driving the driving pile; a data processing device for calculating a penetration amount and a rebound amount of the driving pile by collecting displacement amount image data of the shooting target measured by the camera and rebound amount data measured by the acceleration sensor; and a display device for outputting the penetration amount and rebound amount of the driving pile calculated from the data processing device, wherein the acceleration sensor measures the vibration acceleration value, and the acceleration and displacement algorithm and FFT (Fast Fourier It is an invention using a camera and an acceleration sensor, characterized in that the rebound amount is calculated from the vibration displacement (D) value for each frequency as a single amplitude value by the Transform) process.

However, the prior art also does not reach a satisfactory level in the accuracy of displacement measurement, and the data processing device is inconvenient to use in the field in real time, so field applicability is poor.

Republic of Korea Patent Registration No. 10-2200824 (published on January 12, 2021, title of invention: real-time pile drive penetration and rebound automatic measurement system using camera and acceleration sensor)

The present invention has been devised to solve the above problems, and an object of the present invention is to include both a hammer of a pile driver and two different types of targets attached to a pile in one frame of an image acquired by camera shooting. Not only can the penetration amount, rebound amount, and hammer fall height be measured simultaneously with the video of It is to provide an image-based displacement measuring device related to pile driving.

In addition, since the operator can check the overall operation instructions in the field through the field terminal and directly through wireless communication, the efficiency of driving work and field applicability are excellent, and the occurrence of safety accidents of workers can be prevented. It is to provide a driving related displacement measuring device.

In order to achieve the above object, the present invention is attached to a predetermined position of the hammer and pile of the pile driver, respectively, and two targets having different shapes so as to be distinguished; A camera installed toward the target, continuously acquiring video images to measure the displacement of the target, and including two targets having different shapes in one frame of the acquired video images, and video image data from the camera. It is characterized in that it comprises a field terminal capable of receiving, calculating the amount of displacement of the target, and giving operation instructions by the operator through wireless communication.

In addition, in the present invention, a control unit for performing control related to displacement measurement of the target by transmitting and receiving control signals to and from the camera and the field terminal is included in the field terminal, and the top of the pile is covered with a buffer interior material cap.

In addition, in the present invention, a target detection module for detecting a target from the reference image through moment-based target detection using a reference image, which is a video image obtained by shooting a driving camera, as input data is included in the on-site terminal.

(여기서, n은 타깃이미지의 총픽셀수, x는 타깃이미지와 그에 대응하는 측정이미지(또는 기준이미지)의 가로축 픽셀위치, y는 타깃이미지와 그에 대응하는 측정이미지(또는 기준이미지)의 세로축 픽셀위치, t(x,y)는 타깃이미지의 x,y위치에서의 그레이스케일, m(x,y)는 타깃이미지에 대응하는 측정이미지(또는 기준이미지)의 x,y위치에서의 그레이스케일, σ_t와 σ_m은 각각 타깃이미지와 그에 대응하는 측정이미지(또는 기준이미지)의 그레이스케일의 표준편차,

와

는 각각 타깃이미지와 그에 대응하는 측정이미지(또는 기준이미지)의 그레이스케일의 평균)을 이용하여 계산하는 상관값 계산 모듈은 상기 현장단말기에 포함된다.In addition, the correlation value (value between -1 and +1) for the target image collected by detecting the target in the present invention and the measurement image (or reference image), which is all frames acquired by camera shooting at the time of driving, is the following equation,

(Where n is the total number of pixels of the target image, x is the horizontal axis pixel position of the target image and its corresponding measurement image (or reference image), y is the vertical axis pixel of the target image and its corresponding measurement image (or reference image) position, t(x,y) is the gray scale at the x,y position of the target image, m(x,y) is the gray scale at the x,y position of the measurement image (or reference image) corresponding to the target image, σ _t and σ _m are the standard deviation of the gray scale of the target image and the corresponding measurement image (or reference image), respectively.

and

is an average of gray scales of a target image and a corresponding measured image (or a reference image) respectively). A correlation value calculation module is included in the on-site terminal.

In the present invention, the target image calculates the correlation value for all pixels of the measurement image (or reference image) while moving all pixels of the measurement image (or reference image) by one pixel from the upper left corner to the lower right corner, The correlation value is given to a measurement point (reference point in a reference image), which is a pixel at a constant position among all pixels of a target image, and the measurement point (reference point in a reference image) moves through all pixels of the measurement image (or reference image).

(여기서, α,β,γ는 가장 1에 가까운 위치의 측정이미지(또는 기준이미지) 픽셀과 가로축 세로축으로 각각 인접하는 2개씩의 픽셀에서의 타깃이미지와 측정이미지(또는 기준이미지)에 대한 상관값)을 통해 찾아내는 측정점(또는 기준점) 보간 모듈은 상기 현장단말기에 포함된다.In addition, in the present invention, the correlation value of the target image and the measurement image (or reference image) in the pixel of the measured image (or reference image) at the position where the correlation value is closest to 1 and the two pixels adjacent to each other along the horizontal and vertical axes, respectively. The position of the interpolated measurement point (or interpolated reference point) having the maximum correlation value of the measurement point (or reference point) is determined on the horizontal axis and the vertical axis using the following equations,

(Here, α, β, γ are the correlation values between the target image and the measurement image (or reference image) at the measurement image (or reference image) pixel at the position closest to 1 and two adjacent pixels on the horizontal and vertical axes, respectively. ) The measurement point (or reference point) interpolation module found through is included in the field terminal.

In addition, in the present invention, the interpolated measurement point and the interpolated reference point are used as input data, and the penetration amount, which is the vertical displacement of the position of the interpolated measurement point and the position of the interpolated reference point of the target attached to the pile, and the basis of one driving The rebound amount, which is the vertical displacement of the position of the lowest interpolated measurement point of the target attached to the pile before rebounding and the position of the highest interpolated measurement point of the target attached to the pile after rebounding, and the hammer is at the top of the pile Fall height, which is the vertical displacement of the position of the interpolated measuring point of the target attached to the hammer in a stationary state that is not spaced apart and the position of the interpolated measuring point of the target attached to the hammer in the state in which the hammer is lifted to the maximum, is simultaneously calculated A target displacement calculation module is included in the field terminal.

As described above, the present invention's video-based pile driving-related displacement measuring device includes both the hammer of the pile driver and the two different types of targets attached to the pile in one frame of the image acquired by camera shooting, so that it penetrates into one image Not only can the amount, rebound amount, and hammer fall height be measured at the same time, but the precision of the measured displacement amount is very high by applying the unique correlation value calculation method and interpolation method without being affected by vibration caused by driving, and overall operation in the field Since the operator can check the instructions with the field terminal and directly through wireless communication, the efficiency and field applicability of the driving work are excellent, and the safety of construction can be achieved at the same time.

1 is a view schematically showing a pile driver and a pile equipped with a drop hammer, which is an embodiment used in the present invention.
2 is a view showing a configuration for measuring displacement related to driving in an image-based displacement measuring device related to pile driving according to the present invention.
3 is a diagram schematically illustrating a video image in which two targets of different shapes are photographed in one frame according to the present invention;
4 is a diagram showing an embodiment of a target image, a reference image, and a measurement image in the present invention.
5 is a conceptual diagram illustrating a process of calculating a correlation value between a target image and a reference image (or measurement image) in the present invention;
6 is a conceptual diagram illustrating a parabolic interpolation method applied to the present invention;
7 is a view briefly showing a graph (based on one driving) of measured displacement values analyzed by the measuring device according to the present invention when driving with a hammer.

A preferred embodiment of the present invention configured as described above will be described in detail with reference to the accompanying drawings. Note that the accompanying drawings and the description referring thereto are illustrated so that those skilled in the art can easily understand the present invention, and are not intended to limit the spirit and scope of the present invention. will have to

As shown in FIGS. 1 to 7, the image-based pile driving-related displacement measuring device according to the present invention includes an image collector attached to a cradle, a controller 60, a field terminal 20, and a hammer 50 of a pile driver. It includes targets 30 and 31 attached to a certain position of the pile 40, respectively. The cradle is, for example, a tripod, and the image collector transmits continuously acquired video image data to the field terminal 20 through wired/wireless communication to measure the displacement of the file 40. For example, the target 30, 31), and the hammer 50 is, for example, a drop hammer. A laser collimator may be additionally applied to quickly aim the camera at the targets 30 and 31 .

The control unit 60 transmits and receives control signals to and from the camera 10 and the field terminal 20 to perform a series of overall controls related to displacement measurement, and may be included in the field terminal 20.

The on-site terminal 20 receives video image data from the camera 10 and checks overall operation instructions related to the displacement measurement according to the present invention with the on-site terminal 20, and the operator communicates wirelessly (Bluetooth, Wi-Fi, etc.) ) can be done through Data received by the on-site terminal 20 may be transmitted to a remote integration server.

2 and 3, in the present invention, the targets 30 and 31 are attached to the hammer 50 and the pile 40 of the pile driver, respectively, and the targets 30 and 31 are hit and vibrated by driving. Edo targets (30, 31) must be firmly attached so as not to be moved at the initial attachment point, and the targets (30, 31) can have various shapes, but the targets (30, 31) should be in different forms to be distinguishable.

For example, in the present invention, for easy detection in the video image of the camera 10, a target having a cross shape (+) or scissors shape (×) of the same thickness inside a square border of a certain size of a thick black line ( 30 and 31), and as described above, different internal shapes are used for the hammer 50 and the pile 40 of the pile driver. As will be described later, the targets 30 and 31 of this shape can distinguish different internal shapes using moment-based target detection, and the detection of the targets 30 and 31 and the displacement measurement by tracking according to the position change of the targets An image processed in gray scale (gray scale) is used.

On the other hand, one frame of the image obtained by camera shooting includes both the hammer 50 of the pile driver and the two different types of targets 30 and 31 attached to the pile 40. Here, since the size of the frame is affected by the separation distance between the camera 10 and the pile driver and the angle of view, the distance between the camera 10 and the targets 30 and 31 is determined in consideration of the resolution of the camera and the safety of the operator.

Hereinafter, the process of measuring displacement (intrusion, falling, rebound, etc.) by the displacement measuring device according to the present invention will be described in detail.

First, the targets 30 and 31 are detected from the image acquired by the camera 10, and this detection is performed for both the hammer 50 of the pile driver and the targets 30 and 31 attached to the pile 40. This detection process is a step of acquiring the position and size of the targets 30 and 31 for displacement measurement. That is, the initial positions of the targets 30 and 31 are obtained from the reference image, which is a video image acquired by shooting the driving camera. Depending on changes in lighting or weather, the brightness of the images of the targets 30 and 31 may differ (compensation is possible later), and the size ratio of the images may change depending on the distance between the camera 10 and the targets 30 and 31. there is.

In the present invention, moment-based target detection is applied when detecting the targets 30 and 31. The moments are a series of real values representing the shape of an image, and three functions of Geometric moments, Central moments, and Normalized central moments are used. However, if the central moments are used, the same object can be extracted even if the location of the object is changed, and the normalized central moments are normalized central moments. In addition, there is also the 7 invariant moment method of HU, which has the characteristic that the moment value is invariant for the size, rotation, movement, and symmetry of the image.

This moment-based object detection is a method of comparing shapes using two outlines and grayscale images. The shape information of a certain image is obtained by calculating the correlation between a specific function and an input image. The target, which is an embodiment of the present invention, is a rectangular border of a certain size of a thick black line, and a cross shape (+) of the same thickness is inside it. Or, the detected shape information of a target having a scissors shape (×) is recognized through the controller 60, which is one component of the present invention. The shape information of the other two targets 30 and 31 each have a high correlation value for the same shape information when calculating a correlation value to be described later.

On the other hand, when the targets 30 and 31 are detected by the camera 10, the detection signal is transmitted to the field terminal 20 through the control unit 60, and then tracking and measuring to be described later by the operation instruction by the field terminal 20. steps are performed.

Next, the changed positions of the targets 30 and 31 are tracked and measured from the measurement images, which are all frames acquired by shooting the camera 10 when driving, of the targets 30 and 31 based on the target images collected in the target detection step. Calculate the amount of change in position.

In the present invention, based on the detected target image, the location of the targets 30 and 31 is found in the video image obtained by shooting the camera 10, and the target is obtained from the reference image and the measurement image by using a correlation value calculation method and interpolation method to be described later. It is possible to measure minute positional changes of (30,31). Here, the amount of change in position, that is, the amount of movement of the targets 30 and 31 is calculated in units of mm by converting the pixel size of the image into mm. 4 shows an embodiment of the target image (a), reference image (b), and measurement image (c), and only one target 30 is represented for convenience.

Therefore, in the present invention, the fall of the hammer 50 is basically calculated using the target 31 attached to the pile driving hammer 50, and the amount of penetration and rebound amount using the target 30 attached to the pile 40 yields Since the fall height of the hammer 50 is not a relatively important value compared to the amount of penetration and rebound, the interpolation method described later may not be used, and to prevent damage to the pile 40, a cap containing a buffer material covered on top of the pile 40 ( 41) The error caused by the fluctuation of the level of the upper surface of the cap 41 due to the repeated driving of the hammer 50 is negligible, and the hammer 50 of the pile driver 50 is at the top of the pile 40 every time the hammer 50 is driven. Comparing the position of the target 31 attached to the hammer 50 in a stationary state that is not spaced apart from the position of the target 31 attached to the hammer 50 in the state in which the hammer 50 is lifted to the maximum, the drop height is can be calculated

Table 1 below shows an example in which the amount of penetration, fall and rebound of the pile 40 for each number of driving measured by the measuring device according to the present invention is recorded.

number of hits Measures
(Unit: m) displacement
(Amount of penetration per drive) 5 hits
mean displacement rebound amount fall 0 0 One 0.025 0.025 0.013 1.520 2 0.046 0.021 0.010 1.477 3 0.063 0.017 0.010 1.530 4 0.078 0.015 0.008 1.625 5 0.088 0.010 0.017 0.006 1.550 6 0.098 0.010 0.005 1.456 7 0.106 0.008 0.005 1.448 8 0.111 0.005 0.003 1.510 9 0.116 0.005 0.003 1.580 10 0.120 0.004 0.006 0.002 1.525 .
.
.
.

In Table 1, the measured value is the position difference (vertical displacement) in the vertical axis direction between the last interpolated measurement point (see FIG. 7) and the interpolated reference point, which will be described later after each driving using the target 30 attached to the pile 40 is recorded, and the amount of penetration for each driving by each driving is recorded for the displacement using the measured value, and the rebound amount of the pile 40 generated in the process of driving the pile 40 is also the target attached to the pile 40. (30) is used to calculate and record the interpolated measurement points for all frames during each driving as a standard (see FIG. 7), and the drop will be described later during each driving using the target 31 attached to the pile driving hammer 50. An interpolation method to be described later may also be used to calculate and record the correlation value to be performed.

Specifically, as shown in FIG. 7, each point shown in FIG. 7 on the basis of one-time driving is a measurement point to which the correlation value and interpolation method described later are applied for each frame of the image, and the last point at the viewpoint is the last point interpolated on the basis of one-time driving. It becomes the measurement point and becomes the standard for calculating the amount of penetration, and the vertical displacement difference between the point at the lowest position among the points shown in FIG. 7 and the point at the highest position after rebound becomes the rebound amount.

On the other hand, the average displacement per 5 driving times in Table 1 is the average value of the displacement of the previous 5 driving times after driving 5 times, and as a construction management standard, 'the average displacement per 5 driving times is within the standard range and the average displacement standard value for each 5 driving times' It can be decided as 'to end the driving operation when the range is repeated 3 times'. This reference value may be based on the result of the dynamic load test performed on some piles 40 at the driving site.

Therefore, it is very important to accurately measure the final penetration amount of the pile 40 during driving, and it is also very important to accurately measure the penetration amount according to each driving analysis during driving analysis. Therefore, the present invention can calculate the moving amount (intrusion amount, falling amount, rebound amount, etc.) of the targets 30 and 31 in a unit smaller than a pixel, which is the smallest unit constituting a video image, by applying the correlation value calculation method and the interpolation method. Its precision is very high, so it can achieve the efficiency of driving work and the safety of construction at the same time.

Hereinafter, the correlation value calculation method and the application of the interpolation method in the present invention will be described in detail.

The correlation value between the target image and the measurement image (or reference image) shown in FIG. 5 is calculated as in Equation 1 below.

Here, n is the total number of pixels in the target image, x is the horizontal axis pixel position of the target image and its corresponding measurement image (or reference image), and y is the vertical axis pixel position of the target image and its corresponding measurement image (or reference image) , t(x,y) is the gray scale at the x,y position of the target image, m(x,y) is the gray scale at the x,y position of the measurement image (or reference image) corresponding to the target image.

The gray scale refers to an image in which the value of each pixel is one sample indicating the amount of light, and transmits only light intensity information. This kind of image, also known as black and white or monochromatic, consists of shades of gray, ranging from 'black' at the lightest intensity to 'white' at the highest light intensity. Normally, in 8-bit grayscale, black is 0 and white is 255, so grayscale is expressed as an integer between 0 and 255.

The measurement images are all frames acquired by shooting the camera 10 during driving (see FIG. 4c). For example, if 5 frames are taken per second, 50 measurement images will be generated in the case of 10 seconds, and the above-described correlation for each 50 measurement images. The method of calculating the value is applied. Here, the number of frames per second may be set in consideration of the driving interval.

On the other hand, the reference image is an image obtained by shooting the driving camera 10 (see FIG. 4b), and the method for calculating the correlation value for the target image and the reference image is also replaced with the reference image for the measurement image in the above-described correlation value calculation method. Equation 1 above is applied. That is, if b(x,y) is the gray scale at the x,y position of the reference image corresponding to the target image, m(x,y) is replaced with b(x,y).

For example, if the target image is 10 × 10 pixels, the total number of pixels n is 100, the x is 1 to 10, and the y is 1 to 10, so the measurement image (or reference image) corresponding to the target image has the same size as the target image. becomes 10×10 pixels that are part of the same measurement image (or reference image) (expressed as 5×5 pixels in FIG. 5).

The target image calculates the correlation value for all pixels of the measurement image (or reference image) while moving all pixels of the measurement image (or reference image) preferably by one pixel from the upper left corner to the lower right corner. In other words, if the correlation value is a measurement point (reference point in the reference image) of a pixel at a constant position to which the correlation value is given among all pixels of the target image, the measurement point (reference point in the reference image) is all pixels of the measurement image (or reference image) (In FIG. 5, the part where the target image moves and overlaps with the measured image (or reference image) is expressed in black).

Since it is preferable to set the measurement point (or reference point) at the center pixel or a pixel close to the center of the target image, the measurement point (or reference point) moves through all pixels of the measurement image (or reference image). When calculating the correlation value at the edge of , some pixels of the target image may deviate from the measurement image.

If each image in Equation 1 is divided by the standard deviation and normalized by compensating for the gray scale difference, the following Equation 2 is obtained.

Here, σ _t and σ _m are standard deviations of the gray scale of the target image and the corresponding measurement image (or reference image), respectively.

Equation 2 above has a value between -1 and +1.

In addition, if it is desired to obtain a correlation value that compensates for the difference in image brightness (normalized so that the average is 0 and the standard deviation is 1), it is the same as the following Equation 3, which also has a value between -1 and +1.

와

는 각각 타깃이미지와 그에 대응하는 측정이미지(또는 기준이미지)의 그레이스케일의 평균이다.here,

and

Is the average of the gray scales of the target image and the corresponding measurement image (or reference image), respectively.

Eventually, a correlation value between the target image and the measurement image (or reference image) is calculated using Equation 3, and the position of the target image in the measurement image (or reference image) is found using an interpolation method to be described later. As described above, the target image and the measurement image (or reference image) are converted to gray scale (0 to 255) in advance.

Specifically, among all pixels of the measurement image (or reference image) corresponding to the correlation value of the measurement point (or reference point), a pixel of the measurement image (or reference image) having a correlation value closest to 1 may be found.

Using the interpolation method and the correlation value between the target image and the measurement image (or reference image) at the measurement image (or reference image) pixel at the position closest to 1 and two adjacent pixels on the horizontal axis and vertical axis, respectively, the measurement point ( Or, if the position where the correlation value of the reference point is the maximum (peak, hereinafter referred to as an interpolated measurement point (or interpolated reference point)) is found on the horizontal axis and vertical axis, respectively, the position of the interpolated measurement point (or interpolated reference point) is the target ( 30, 31) is used as a standard for calculating the displacement (intrusion, falling, rebound, etc.).

In other words, the interpolated measurement point (or interpolated reference point) is a measurement point (or reference point) at which the correlation value of the measurement point (or reference point) is estimated as the maximum value, and the concept of the interpolation method is as shown in FIG. 6, and a parabola The general formula for is shown in Equation 4 below.

α, β, γ (the measurement image (or reference image) pixel at the position closest to 1), which is the three correlation values closest to the maximum value (peak) of the measurement point (or reference point), and two adjacent on the horizontal axis and vertical axis, respectively. If the correlation value of the target image and the measurement image (or reference image) at the pixel of is used, the position of the interpolated measurement point (or interpolated reference point) is as shown in Equation 5 below.

As described above, the positional difference between the interpolated measurement point and the interpolated reference point becomes the displacement of the targets 30 and 31, that is, the amount of displacement in the vertical axis direction (vertical displacement) corresponds to the amount of penetration of the pile 40 per frame and falling and rebound. quantity, etc.

On the other hand, in the present invention, calculation of average displacement (see Table 1) for n times per stroke, detection of targets 30 and 31, calculation of correlation values, interpolation of measurement points (or reference points) and displacement of targets 30 and 31 (penetration A program coded as an algorithm through a programming language is used to perform these processes with a computer when calculating the amount of weight, fall, rebound, etc.).

Accordingly, the average displacement calculation module calculates the average value of the displacement of n times immediately before driving after n times of driving with the amount of penetration for each driving as input data. In other words, this program is stored in the field terminal 20 or a separate storage device, so that the average displacement calculation module uses the input data stored in the field terminal 20 or a separate storage device and the program to find displacement n times. will calculate the average value of .

The target detection module detects the targets 30 and 31 from the reference image through moment-based target detection using the reference image obtained before driving as input data. This detection process is finally performed by using a program language In other words, this program includes moment-based object detection and is stored in the field terminal 20 or a separate storage device so that the target detection module is input and stored in the field terminal 20 or a separate storage device. The targets 30 and 31 are detected using the input data and the program.

The correlation value calculation module calculates a correlation value that satisfies Equation 3 using Equation 3 and the input data. This calculation process is a program coded as an algorithm through a programming language to be finally performed by a computer, In other words, this program includes Equation 3 and is stored in the field terminal 20 or a separate storage device, so that the correlation value calculation module converts the input data stored in the field terminal 20 or a separate storage device and the program. It will be used to calculate the correlation value.

The measurement point (or reference point) interpolation module determines an interpolated measurement point (or interpolated reference point) that satisfies Equation 5 using Equation 5 and the input data. It is a program coded with an algorithm through, in other words, this program includes Equation 5 and is stored in the field terminal 20 or a separate storage device so that the measurement point (or reference point) interpolation module is the field terminal 20 or a separate storage device. An interpolated measurement point (or an interpolated reference point) is determined using the input data stored in the storage device and the above program.

As described above, the displacements of the targets 30 and 31 required in the present invention are the penetration amount, the rebound amount, and the falling height. The penetration amount is the vertical displacement of the interpolated measurement point position and the interpolated reference point position, and the rebound amount is one time. It is the vertical displacement of the position of the lowest, interpolated measurement point before rebound on a driving basis and the position of the highest, interpolated measurement point after rebound, and the hammer 50 is not spaced apart from the top of the pile 40 after falling. Vertical of the position of the interpolated measuring point of the target 31 attached to the hammer 50 at , and the position of the interpolated measuring point of the target 31 attached to the hammer 50 in the state in which the hammer 50 is lifted to its highest. is the displacement

Accordingly, the target displacement calculation module calculates the displacement of the targets 30 and 31 using the interpolated measurement point (or interpolated reference point) as input data. In other words, this program is stored in the field terminal 20 or a separate storage device, and the target displacement calculation module uses the input data stored in the field terminal 20 or a separate storage device and the program The displacement of the targets 30 and 31 is calculated.

The average displacement calculation module, target detection module, correlation value calculation module, measurement point (or reference point) interpolation module, and target displacement calculation module are included in the field terminal 20, and the modules allow the present invention to be performed on a computer. As technical means for this, an average displacement calculation unit, a target detection unit, a correlation value calculation unit, a measurement point (or reference point) interpolation unit, and a target displacement calculation unit may be respectively named.

Therefore, the present invention utilizes a camera with a lens having a magnification suitable for the field to easily search for a target with a wide angle of view, and the existing line scan method, which is difficult to align in field setup, and the QR code method, which must be measured at a short distance due to a weak optical base. Compared to this, it has excellent field applicability, and by applying the correlation value calculation method and interpolation method, the amount of movement of the target (intrusion amount, fall, rebound amount, etc.) can be calculated in units smaller than pixels, the smallest unit constituting a video image. High precision.

10: camera 20: field terminal
30: target attached to pile 31: target attached to hammer
40: pile (pile) 41: buffer material built-in cap
50: hammer 60: control unit

Claims (7)

two targets 30 and 31 each attached to a predetermined position of the hammer 50 and the pile 40 of the pile driver and having different shapes so as to be distinguished;
It is installed toward the targets 30 and 31 and continuously acquires video images to measure the displacement of the targets 30 and 31, and in one frame of the acquired video images, the two targets 30 with different shapes 31), and
It includes a field terminal 20 capable of receiving video image data from the camera 10, calculating the amount of displacement of the targets 30 and 31, and giving operating instructions by an operator through wireless communication,
A target detection module for detecting targets 30 and 31 from the reference image through moment-based target detection using the reference image, which is a video image acquired by shooting the driving camera 10, as input data Included in the field terminal 20 become,
The correlation value (value between -1 and +1) between the target image collected by detecting the target 30 and 31 and the measurement image (or reference image), which is all frames acquired by shooting the camera 10 at the time of driving, is as follows math formula,

(Where n is the total number of pixels of the target image, x is the horizontal axis pixel position of the target image and its corresponding measurement image (or reference image), y is the vertical axis pixel of the target image and its corresponding measurement image (or reference image) position, t(x,y) is the gray scale at the x,y position of the target image, m(x,y) is the gray scale at the x,y position of the measurement image (or reference image) corresponding to the target image, σ _t and σ _m are the standard deviation of the gray scale of the target image and the corresponding measurement image (or reference image), respectively.

and

is the average of the gray scales of the target image and the corresponding measured image (or reference image) respectively), characterized in that the correlation value calculation module is included in the field terminal 20, image-based file search Associated displacement measuring device.
According to claim 1,
A control unit 60 for performing control related to displacement measurement of the targets 30 and 31 by transmitting and receiving control signals with the camera 10 and the field terminal 20 is included in the field terminal 20, and the file ( 40) An image-based pile driving-related displacement measuring device, characterized in that the cap 41 of the buffer interior material is covered at the top.
delete delete According to claim 1,
The target image calculates the correlation value for all pixels of the measurement image (or reference image) while moving all pixels of the measurement image (or reference image) by one pixel from the upper left corner to the lower right corner. It is given to a measurement point (reference point in the reference image), which is a pixel at a certain position among all pixels in the image, and the measurement point (reference point in the reference image) moves through all pixels of the measurement image (or reference image), image-based Displacement measuring device related to pile driving.
According to claim 5,
The measurement point is obtained by using the correlation values of the target image and the measurement image (or reference image) at the pixel of the measurement image (or reference image) at the position where the correlation value is closest to 1 and the two adjacent pixels on the horizontal and vertical axes, respectively. The position of the interpolated measurement point (or interpolated reference point) at which the correlation value of (or reference point) is the maximum is determined by the following equations on the horizontal axis and the vertical axis, respectively;

(Here, α, β, γ are the correlation values between the target image and the measurement image (or reference image) at the measurement image (or reference image) pixel at the position closest to 1 and two adjacent pixels on the horizontal and vertical axes, respectively. ) The measuring point (or reference point) interpolation module found through is characterized in that it is included in the field terminal 20, the image-based displacement measuring device related to pile driving.
According to claim 6,
Using the interpolated measurement point and the interpolated reference point as input data,
The amount of penetration, which is the vertical displacement of the position of the interpolated measurement point and the position of the interpolated reference point of the target 30 attached to the pile 40,
Interpolation of the position of the interpolated measuring point of the target 30 attached to the pile 40, the lowest before rebound on the basis of one driving, and the position of the target 30 attached to the pile 40, the highest after rebound The rebound amount, which is the vertical displacement of the position of the measured point, and
The position of the interpolated measurement point of the target 31 attached to the hammer 50 in a stationary state in which the hammer 50 is not spaced apart from the top of the pile 40 and the hammer in a state in which the hammer 50 is lifted to the highest level. A target displacement calculation module that simultaneously calculates the vertical displacement of the position of the interpolated measurement point of the target 31 attached to (50) is included in the field terminal 20, image-based file driving. Associated displacement measuring device.

Images